Android app missing large chunks of timeline, with errors
Using the Android app, when loading the timeline, large chunks (multiple years, hundreds of photos) are missing. Smart search in the app is behaving oddly too (sometimes not showing any photos when filtering by person, location, etc). Everything appears perfectly through the web interface without error, even when logging in for the first time with incognito on a browser never before used with Immich.
App reinstalled, docker pulled to current release.
Primary client app error (appears every time app is opened):
Server error appearing at the same time as above:
Also, while building the timeline, there are a few (<10) client and server errors related to missing thumbnails. I have tried regenerating all thumbnails but this has not resolved the issue. Example server error is below:
Thank you so much in advance!!
44 Replies
:wave: Hey @ZanyHunter,
Thanks for reaching out to us. Please carefully read this message and follow the recommended actions. This will help us be more effective in our support effort and leave more time for building Immich :immich:.
References
- Container Logs:
docker compose logs docs
- Container Status: docker ps -a docs
- Reverse Proxy: https://immich.app/docs/administration/reverse-proxy
- Code Formatting https://support.discord.com/hc/en-us/articles/210298617-Markdown-Text-101-Chat-Formatting-Bold-Italic-Underline#h_01GY0DAKGXDEHE263BCAYEGFJA
Checklist
I have...
1. :ballot_box_with_check: verified I'm on the latest release(note that mobile app releases may take some time).
2. :ballot_box_with_check: read applicable release notes.
3. :ballot_box_with_check: reviewed the FAQs for known issues.
4. :ballot_box_with_check: reviewed Github for known issues.
5. :ballot_box_with_check: tried accessing Immich via local ip (without a custom reverse proxy).
6. :ballot_box_with_check: uploaded the relevant information (see below).
7. :ballot_box_with_check: tried an incognito window, disabled extensions, cleared mobile app cache, logged out and back in, different browsers, etc. as applicable
(an item can be marked as "complete" by reacting with the appropriate number)
Information
In order to be able to effectively help you, we need you to provide clear information to show what the problem is. The exact details needed vary per case, but here is a list of things to consider:
- Your docker-compose.yml and .env files.
- Logs from all the containers and their status (see above).
- All the troubleshooting steps you've tried so far.
- Any recent changes you've made to Immich or your system.
- Details about your system (both software/OS and hardware).
- Details about your storage (filesystems, type of disks, output of commands like fdisk -l and df -h).
- The version of the Immich server, mobile app, and other relevant pieces.
- Any other information that you think might be relevant.
Please paste files and logs with proper code formatting, and especially avoid blurry screenshots.
Without the right information we can't work out what the problem is. Help us help you ;)
If this ticket can be closed you can use the /close command, and re-open it later if needed.Have you been moving the database or folders around?
I have not touched the backend folders since my initial deployment of the service in mid-2024.
docker-compose.yml, .env content
I think your database might be corrupted
can you post the postgres logs? they might be in DB_DATA_LOCATION
What is the server version?
Logs from last time it was started. I truncated, as the same error and select statement is repeated every time the app is loaded. I'm curious why it is requesting to allocate 2 integers lower than the maximum representable unsigned 64-bit integer (18446744073709551615).
Server: 1.123.0
App: 1.123.0 build.172
can you post
docker ps -aI think the database is healthy
not sure what is causing that large allocation. perhaps a misformed query of some kind
Odd. Glad you suspect the database is healthy. I am waiting on an old iPhone 11 to charge so I can test the iPhone mobile app. Trying to narrow it down.
how much RAM on this system?
Originally 4GB, upped to 8GB yesterday to troubleshoot this issue. 1.75GB currently in use.
tell us more about the host please
sounds like a VM?
Yes, this is an Ubuntu Server 24.04.1 LTS VM running on an Intel-based TrueNAS Core server. The database volume is local, the library is mapped via NFS.
I also have an available Intel-based Proxmox cluster I can migrate to if a more standard virtualization paltform is desired.
When you open up the app to sync, do you see anty error from the server log?
if it's a true VM (which this is, not LXC) I don't think that would present a factor
I saw some similar reports but most are from outdated server
so this case is strange
Server presents this error when the app is opened (copied from OP):
App has an associated 500 error log.
Can you try
docker compose down && docker compose pull && docker compose up?I downed, purged, re-pulled, and upped earlier during initial troubleshooting. Just did it again for good measure. I see the same errors when opening the app again.
Can you try access over local IP?
Same error when connected via IP:port.
On the Android phone (Pixel 7 Pro running Android 15 public release), Immich was uninstalled and reinstalled from the Play Store, and app storage & cache were cleared. Phone was also rebooted.
what is the size of the user's library?
20,100 in timeline, 691 in trash
what %/number of those are on the mobile device?
<2,000 assets
is there any correlation with the missing assets?
I'm having a lot of trouble finding a correlation. Of the ones that are appearing, some are local on this phone, some are not, and some photos that I know are local on this phone are missing from the timeline.
In the mobile timeline, I have photos from present (today) through December 22, 2024, then nothing until a single photo from October 2022, then nothing until February 2022, etc.
When the app was reinstalled, it lost permissions to the local filesystem, so I don't know how local photos would influence what appears on the timeline (correct me if I'm wrong!)
Just tested on an old Samsung Galaxy Note 8. Freshly reset OS, fresh install of the Immich app. The same exact sporadic photos on the same dates appear on it as well, with the same error. Will try an iPhone soon.
It goes without saying that there are no local photos on this device.
Hmm I wonder if you can try a different virtualization environment
Here's what I did:
Installed Ubuntu 24.04.1 in a Proxmox VM
apt update, apt upgrade
Installed Docker using docs.docker.com's install guide
Copied docker-compose.yml, .env, and postgres/ to new VM
Mounted library NFS on new VM
Started Immich
Connected with app via direct IP
Same errors, same symptoms. Same random photos missing from the timeline on Android. Web seems completely fine?
I'd obviously rather not start from scratch. Do you have any other suggestions before I go nuclear on the database? I would rather not have to rename faces, recreate user accounts, re-share, etc.
Ohhh, I got something. I tried scrolling to August 2023 on the web, and got a 500 error. Docker logs were flooded with dozens of variations of this error. Pulling postgres logs now.
Here's the related postgres logs
I think your database is toast
Was it stored on a LXC mount?
Is your database even dumpable? I suspect not
Yeah, I suspected that unfortunately. It was not on an LXC mount. I suspect perhaps there was an unclean shutdown at some point or something. What I ended up doing was mounting a snapshot I had of the VM from 2 weeks ago. That was not old enough, so I restored the oldest internal Immich backup from that snapshot. Looks like I'm good to go from that backup! Now I just need to figure out how to remove references to images uploaded and deleted since the restore (lots of "error loading image" on the screen). Let me attempt a dump of the corrupted database and see what it does just for posterity.
You can use the unlisted /admin/repair page to find the untracked files
Is your postgres container healthy?
Im surprised a hard shutdown would kill it but odds are proxmox didnt properly flush a write or theres a storage device issue
I have 2 VMs running, one is restored from the past, one is current and corrupted. On the corrupted one, it's still healthy even after dying when trying to load August 2023.
My VMs run on a raidz1 zpool, and I don't seen any pool inconsistencies or errors. Might never know what happened.
Here's the result of trying to dump the DB. Same error!
An edge case, very hard to research. GPT seems to think that the handling of sync writes on LXC is not as reliable vs a VM
So if ZFS didnt issue back a sync but postgres was told it was, that can cause an issue.
Definitely. Makes sense.
I'm effectively back up and running. Just waiting on the Repair page. Massive thank you! If I can return the favor or buy you and @Alex a coffee, let me know.
No worries. You can get a license when able!
Absolutely! Will do.
This thread has been closed. To re-open, use the button below.